PROJECT SUMMARY (See instructions); Understanding the complex relationship between obesity and cancer is a scientific priority with significant public health implications in reducing the wide-spread burden of these two common chronic diseases. The goals of Project 2 are to: (1) identify genetic variants that influence high-risk obesity phenotypes; (2) uncover the relationship between host genes, gut microbiota, and obesity; (3) determine the association between genetic ancestry and obesity phenotypes, usefully providing a biological measure of race/ethnicity to all four projects; and (d) help shed light on the key exposure predictors that are most relevant for obesity, breast and colorectal cancers. We propose in Aim 1 to conduct a genome-wide association study (GWAS) of body fat distribution. Specifically, we will genotype over 4.3 million single nucleotide polymorphisms in 2,000 Multiethnic Cohort participants, common to all four projects, who will undergo body composition assessment by whole-body dual energy X-ray absorptiometry and abdominal magnetic resonance imaging. Next in Aim 2, we will conduct a genome-wide association study of the relative abundance of the gut microbiota and, more specifically, the gut microbial profiles associated with obesity. Here we will combine genotyping data from the 2,000 MEC subjects examined in Aim 1 with existing GWAS data from over 4,200 MEC participants. For these 6,200 subjects, we will characterize the architecture of the gut microbiota by 16S gene rRNA sequencing of stool samples in Project 4. To our knowledge, this will be the first large-scale comprehensive investigation of the relationship between host genetics and the gut microbial community, and in particular gut microbial profiles associated with obesity. Lastly, our final aim will integrate genetic results from Project 2 with those from the other projects to study the interrelationships across data dimensions (diet, lifestyle/behaviors, biochemical and hormonal factors, metabolites, and gut microbial composition), construct best predictive models of body fat amount and distribution, and explore the relation of the predicted values with breast and colorectal cancer risks. This last aim will include testing risk variants for obesity phenotypes and gut microbial composition with breast and colorectal cancer susceptibility. Association testing will be conducted in our large case-control studies of breast (3,243 cases, 3,243 controls) and colorectal (2,588 cases, 2,588 controls) cancer nested within the Multiethnic Cohort. The strengths of this proposal include: 1) the innovativeness of the research; 2) the multi-disciplinary investigative team, 3) the efficient use of existing resources and 4) the scientific and public health significance of obesity and cancer. The knowledge gained by this study may significantly advance our understanding of the biological factors driving body fat distribution and their effects on the development of common cancers among several racial/ethnic populations.